Examkin

Question

Difficulty: HardHigh-Performing Database Solutions

A financial technology enterprise is designing a global transaction ledger application. The system must process high-frequency writes (over 100000100{}000 transactions per second at peak) with sub-1010-millisecond latency. Users in North America, Europe, and Asia must be able to query their local transaction history with sub-22-millisecond latency. The database must also support transactional ACID properties across multiple records. Which two solutions should the solutions architect implement to meet these high-performance database requirements? (Select TWO).

  1. Configure Amazon DynamoDB Global Tables with a partition key design that combines a high-cardinality customer identifier with a random suffix to distribute write traffic evenly across physical partitions.Answer
  2. Implement DynamoDB Transactions to execute ACID-compliant updates across multiple items, ensuring consistency and single-digit millisecond performance for ledger updates.Answer
  3. C
    Configure Amazon DynamoDB Global Tables using a partition key based on a monotonically increasing timestamp to ensure chronological sorting of transaction records within each partition.
  4. D
    Deploy Amazon RDS Multi-AZ DB clusters across regions and create Read Replicas, configuring the application to route write traffic to the nearest Read Replica to maintain write availability.

Answer

The correct solutions are to configure Amazon DynamoDB Global Tables using a sharded partition key design that combines a customer identifier with a random suffix, and to implement DynamoDB Transactions to handle ACID-compliant multi-record updates.
The correct configuration uses Amazon DynamoDB Global Tables to achieve active-active multi-region replication with sub-10-millisecond latency. To handle the scale of 100000100{}000 writes per second, the partition key must be designed with high cardinality (such as a customer identifier with a random suffix) to distribute the workload evenly across physical partitions and avoid write bottlenecks. Additionally, DynamoDB Transactions satisfies the transactional ACID requirements with low latency.

Step-by-Step Solution

1
Evaluate the read and write latency requirements against AWS database services.
Amazon DynamoDB Global Tables can deliver sub-10-millisecond write latency and sub-2-millisecond read latency globally by executing operations locally in the nearest AWS region.
Relational options like Amazon RDS across multiple regions introduce high latency for write operations and failovers, which do not meet the performance and scaling goals.
2
Address the high-frequency write scaling strategy to support peak traffic of over 100000100{}000 writes per second.
Implement a partition key strategy that uses a composite design combining a customer ID with a random suffix (write sharding).
In DynamoDB, partition keys determine which physical partition receives the write. Distributing writes evenly across partitions prevents hot partitions and avoids throughput bottlenecks.
3
Determine the mechanism to support ACID transaction properties.
Enable DynamoDB Transactions to execute coordinated read/write operations across multiple items in the transaction ledger.
DynamoDB Transactions guarantees ACID compliance (atomicity, consistency, isolation, and durability) within single-digit millisecond latency envelopes.

Key Concept

Amazon DynamoDB Global Tables combined with write sharding and DynamoDB Transactions allows global, active-active scale with low latency and transaction consistency.
Estimated Time:3m 0s
Rate this question

Topics

High-Performing Database Solutions

More questions from this topic